Justin Reese
Computer Research Scientist
Research Interests
I am broadly interested in using computational methods to extract actionable knowledge from biomedical and biological data. I am particularly interested in using these methods to address human disease.
My current research focuses on developing performant machine learning algorithms to extract knowledge from biomedical data. This involves for example applying unsupervised machine learning on EHR data for identifying subtypes of human diseases such as long COVID, and constructing and performing graph machine learning on knowledge graphs such as KG-COVID-19 (https://github.com/Knowledge-Graph-Hub/kg-covid-19) and using this knowledge to identify drugs that may have an effect on COVID-19 outcome.
I am also developing and applying epidemiological methods to validate knowledge extracted from computational experiments using EHR and other clinical data, for example describing the effect of metformin on COVID-19 outcome.
Recent Publications
Related News
Machine Learning Tackles Long COVID
A new machine learning tool developed by a team of researchers led by Justin Reese of Berkeley Lab and Peter Robinson of Jackson Lab analyzes electronic health records to find symptoms in common between people who have been diagnosed with long COVID and to define subtypes of the condition.
Metformin May Mitigate More Severe COVID-19 Outcomes in Patients with Prediabetes, PCOS
An international team led by Justin Reese, a research scientist in the Environmental Genomics and Systems Biology (EGSB) Division, analyzed electronic health record data aggregated in the National COVID Cohort Collaborative (N3C) Data Enclave to assess whether metformin is associated with reduced COVID-19 severity in people with prediabetes or polycystic ovary syndrome (PCOS), two common conditions that increase the risk of severe COVID-19 presentation.
